Cooperative Object Grasping and Transportation by Multi-agent Systems Considering Grasping Force Constraints

Abstract

In recent years, cooperative object transportation using multi-agent systems has attracted attention due to its flexibility in handling various object sizes and operating in constrained environments. This study proposes a control method that achieves stable transportation while considering grasping constraints and the rotational motion of the object. Existing approaches use fixed reference forces determined in advance, which may lead to insufficient or excessive grasping when the object rotates, compromising stability. In contrast, our method formulates constraint conditions to maintain grasping and derives force relations to suppress rotation. By optimizing the reference force at each time step, each agent can adaptively generate appropriate grasping forces according to the object's current state. Simulation results demonstrate that, even during the object rotates, the proposed method maintains grasping and enables convergence to the desired velocity. Compared to the existing method, our approach exhibits superior transportation performance, confirming its effectiveness for robust and flexible cooperative transport.